XiaomiMiMo-V2.6-Pro
5 cheaper ranked models score higher: DeepSeek V4.1 Flash (58.9, $0.518), DeepSeek V4.1 Flash 1017 (58.9, $0.525), GPT-5.6 Luna (58.4, $0.450) and 2 more.
As of 9 Nov 2026, MiMo-V2.6-Pro by Xiaomi scores 57.1 on the BenchLeader Index (599 ranked configurations), with its best results in long context (68) and weakest in knowledge (42). API pricing: $0.440 per 1M input tokens and $0.870 per 1M output tokens ($0.544 blended at 3:1), list price on 9 Nov 2026. It streams about 51 tokens/s (Artificial Analysis). Released 21 Sep 2026.
Key facts (as of 9 Nov 2026)
- MiMo-V2.6-Pro API pricing: $0.440 per 1M input tokens and $0.870 per 1M output tokens ($0.544 blended at 3:1) (list price, 9 Nov 2026).
- MiMo-V2.6-Pro streams about 51 tokens/s (Artificial Analysis, 9 Nov 2026).
- MiMo-V2.6-Pro was released on 21 Sep 2026 (Xiaomi).
In detail. MiMo-V2.6-Pro is a Xiaomi open-weights reasoning model, released 21 Sep 2026. Its best configuration ranks #157 of 599 on the BenchLeader Index at 57.1 ±3.9, in the upper half. The ± is the point: 249 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in long context (68) and lowest in knowledge (42). At $0.548 per million tokens blended it is cheaper than most ranked models. Output speed of 51 tokens per second puts it slower than most, with a first answer in 25.7 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 8 Nov 2026.
- Blended price
- $0.548/M
- $0.440 in · $0.870 out
- via Artificial Analysis
- Output speed
- 51 tok/s
- measured by Artificial Analysis
- First answer
- 26 s
- first token 2.69 s
- Context
- 1.0M
- via models.dev
- Full answer
- 101 s
- median, reasoning included
- Cost per task
- $0.257
- 39 min a task, AA Index
- Tokens per task
- 121k
- 71% of it reasoning
- Released
- 21 Sep 2026
- Overall index57
- Reasoning50
- Coding59
- Agents & tools58
- Maths66
- Knowledge42
- Human preference67
- Multimodal60
- Long context68
- Composite52
Versions
Xiaomi has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| MiMo-V2.6-Prothis page | 21 Sep 2026 | 57.1 | #157 |
| MiMo-V2.5-Pro | 23 Apr 2026 | 55.6 | #201 |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | – | – | 51 tok/s | 26 s | $0.0004 | |
| not statedbest | 57.1 | #157 | 51 tok/s | 26 s | $0.0004 | Agents & tools 58 · Coding 59 · Composite 52 · Human preference 67 · Knowledge 42 · Long context 68 · Maths 66 · Multimodal 60 · Reasoning 50 |
How its index has moved
25 Sep 2026 to 9 Nov 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Metadata
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA Openness Indexnot in index | 52.8#26 | Artificial Analysis |
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Hard Prompts | 1504#27 | Arena (LMArena) | |
| LiveBench 2.0 Reasoningnot in index | 79.6%#25 | LiveBench | |
| Humanity's Last Exam (AA)not in index | 49.4%#34 | Artificial Analysis | |
| HLE-Diamond (AA) | 15.2%#93 | Artificial Analysis | |
| CritPtnot in index | 26.6%#43 | Artificial Analysis | |
| CritPt v2 hard subset (AA) | 34.0%#63 | Artificial Analysis | |
| MysteryMechanismnot in index | 15.3%#28 | Vals AI |
Coding
| Benchmark | no reasoning | not stated | Source | Trend |
|---|---|---|---|---|
| SciCodeeffort not stated | – | 60.9%#8 | SciCode | |
| Arena Codingeffort not stated | – | 1538#13 | Arena (LMArena) | |
| Arena WebDeveffort not stated | – | 1624#25 | Arena (LMArena) | |
| LiveBench 2.0 Codingnot in indexeffort not stated | – | 64.5%#24 | LiveBench | |
| SciCode (AA)not in indexeffort not stated | – | 60.9%#10 | Artificial Analysis | |
| IOIeffort not stated | – | 39.3%#37 | Vals AI | |
| Code Migrationnot in indexeffort not stated | – | 43.0%#27 | Vals AI | |
| Code Migration v1.1not in indexeffort not stated | – | 48.8%#17 | Vals AI | |
| ProgramBenchnot in indexeffort not stated | – | 0.5%#19 | Vals AI | |
| SWE-bench Multimodalnot in indexeffort not stated | – | 41.3%#17 | Vals AI | |
| Vibe Code Bench v1.1not in indexeffort not stated | – | 85.2%#17 | Vals AI | |
| ALE-Benchnot in indexeffort not stated | – | 1158.1#52 | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in indexeffort not stated | – | 34.9%#50 | Artificial Analysis | |
| GameBench 2 by reasoning settingnot in index | 49.6#88 | – | DuelLab GameBench 2 | |
| Design Arena Websitenot in indexeffort not stated | – | 1305#18 | Design Arena | |
| FrontierCode 1.1 (AA)not in indexeffort not stated | – | 45.4%#20 | Artificial Analysis | |
| FrontierCode 2.0 (AA)effort not stated | – | 24.0%#51 | Artificial Analysis |
Agents & tools
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| ProofBench | 70.0%#18 | Vals AI | |
| Arena Maths | 1482#35 | Arena (LMArena) |
Knowledge
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench 2.0 Data Analysisnot in index | 72.3%#22 | LiveBench | |
| LiveBench Researchnot in index | 84.8%#11 | LiveBench | |
| AA-Omniscience v1.1 | -4.5#87 | Artificial Analysis | |
| Excel Modeling Benchmarknot in index | 62.9%#30 | Vals AI | |
| Excel Modeling Benchmark v1.1not in index | 67.3%#29 | Vals AI | |
| MedCodenot in index | 45.0%#39 | Vals AI | |
| MedScribenot in index | 88.3%#13 | Vals AI | |
| Public Benefits Bench v1.1not in index | 68.9%#8 | Vals AI | |
| AA-Omniscience v1.1: accuracynot in index | 34.8%#79 | Artificial Analysis | |
| AA-Omniscience v1.1: non-hallucinationnot in index | 39.8%#83 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Instruction Followingnot in index | 1472#37 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1480#29 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1448#42 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1463#76 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1497#21 | Arena (LMArena) |
Multimodal
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Vision | 1266#48 | Arena (LMArena) | |
| LiveBench Designnot in index | 74.3%#16 | LiveBench | |
| SAGEnot in index | 84.5%#20 | Vals AI | |
| Code2Video-AAnot in index | 1052#11 | Artificial Analysis | |
| AA-3Dnot in index | 983#16 | Artificial Analysis |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR v1.1 | 86.3%#3 | Artificial Analysis | |
| MLCRnot in index | 12.8%#97 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench 2.0 | 66.1%#16 | LiveBench | |
| AA Intelligence Index v5.0not in index | 29.9#62 | Artificial Analysis | |
| AA Intelligence Index v4.3.2not in index | 46.3#35 | Artificial Analysis | |
| AA Progress Index v1not in index | 321.2#55 | Artificial Analysis | |
| Vals Indexnot in index | 57.8#19 | Vals AI | |
| Vals Finance Indexnot in index | 57.9%#17 | Vals AI | |
| Vals Router Indexnot in index | 57.8%#12 | Vals AI | |
| AA Cyber Indexnot in index | 56.1#4 | Artificial Analysis |
GameBench 2, best reasoning setting (None): 49.6. Reference only, not part of the BenchLeader Index.
DuelLab GameBench 2, release gb2-20261103-eaea31df94e6 (https://benchmarks.duellab.org/releases/gb2-20261103-eaea31df94e6/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
API pricing: where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| PrimaLabs | 140 tok/s | 0.80 s | $4.00 | $8.00 | 1.0M | fp8 |
| GMICloud | 47 tok/s | 3.10 s | $0.326 | $0.652 | 1.1M | bf16 |
| Xiaomi | 40 tok/s | 2.29 s | $0.435 | $0.870 | 1.0M | fp8 |
| StreamLake | 40 tok/s | 1.23 s | $0.522 | $1.04 | 1M | – |
| DeepInfra | 36 tok/s | 0.84 s | $0.387 | $0.783 | 1.0M | fp8 |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.004 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0004 | $0.0003 | 31.5 s |
| Summarize a 30-page report | 12,000 / 600 | $0.0058 | $0.0019 | 37.3 s |
| Code edit | 6,000 / 1,500 | $0.0039 | $0.0020 | 54.9 s |
| Agentic coding session | 60,000 / 4,000 | $0.030 | $0.010 | 1.7 min |
| Structured extraction | 2,000 / 200 | $0.0011 | $0.0004 | 29.6 s |
Compare MiMo-V2.6-Pro with…
See also
Embed or cite
Show where MiMo-V2.6-Pro ranks on a model card or README. The badge redraws itself with every daily update, carries the data date and links back to this page. To see ranks like this on OpenRouter and Hugging Face model pages, get the Chrome extension.
[](https://www.benchleader.com/models/mimo-v2-6-pro)Cite this
The ranking changes daily, so a reference carries the data date. APA style:
BenchLeader. (2026, November 9). MiMo-V2.6-Pro on the BenchLeader Index (method v1.4) [Data set]. Retrieved November 9, 2026, from https://www.benchleader.com/models/mimo-v2-6-pro
BibTeX:
@misc{benchleader_mimo_v2_6_pro_20261109,
author = {{BenchLeader}},
title = {{MiMo-V2.6-Pro on the BenchLeader Index (method v1.4)}},
year = {2026},
month = nov,
howpublished = {\url{https://www.benchleader.com/models/mimo-v2-6-pro}},
note = {Data as of 2026-11-09}
}Every badge style, the embeddable top-10 widget and how to cite the Index
Questions people ask
- What are MiMo-V2.6-Pro's benchmark scores?
- As of 9 Nov 2026, MiMo-V2.6-Pro scores 57.1 ±3.9 on the BenchLeader Index, strongest in long context (68) and weakest in knowledge (42). That rests on independent benchmark results only, and every score on this page links to its publisher.
- How does MiMo-V2.6-Pro rank?
- MiMo-V2.6-Pro is #157 of 599 ranked configurations on 9 Nov 2026. 249 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Fable 5.2 (max effort) at 69.4.
- How much does MiMo-V2.6-Pro cost per 1M tokens?
- As of 9 Nov 2026, MiMo-V2.6-Pro costs $0.440 per 1M input tokens and $0.870 per 1M output tokens, $0.548 blended (three input tokens per output token).
- Is MiMo-V2.6-Pro better than GLM 5.3?
- As of 9 Nov 2026, GLM 5.3 (max) and MiMo-V2.6-Pro are neck and neck on quality (58.9 vs 57.1 on the BenchLeader Index, inside the ±4.9 margin of error), and MiMo-V2.6-Pro costs 3.9× less, which makes it the better buy for most workloads. The head-to-head page has the benchmark-by-benchmark detail.
- When was MiMo-V2.6-Pro released?
- Xiaomi released MiMo-V2.6-Pro on 21 Sep 2026. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 9 Nov 2026. Compare these configurations.
Cite as: BenchLeader, “MiMo-V2.6-Pro: benchmarks, pricing, speed and rank”, data as of 9 Nov 2026, https://www.benchleader.com/models/mimo-v2-6-pro.